Don't have Telegram yet? Try it now!
https://hgpu.org/?p=11038
Using Shared Memory as a Cache in Cellular Automata Water Flow Simulations on GPUs